Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a comparison, exception, or differentiation

    oai_token-act-pair · gpt-3.5-turbo

    comparative phrases that indicate exceptions or alternatives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    orian
    -0.78
    ahime
    -0.73
    fell
    -0.71
    foreseen
    -0.71
    uese
    -0.70
    natureconservancy
    -0.70
    oro
    -0.70
    ãĥł
    -0.69
     Cosponsors
    -0.67
    izer
    -0.67
    POSITIVE LOGITS
     oneself
    0.80
     maybe
    0.79
     those
    0.74
     perhaps
    0.74
     possibly
    0.71
     ours
    0.70
     namely
    0.70
     theirs
    0.70
     ourselves
    0.69
     hers
    0.66
    Act Density 0.054%

    No Known Activations